What Generative AI actually is in Prompt Engineering / GenAI - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding Generative AI Outputs

Which of the following best describes what Generative AI does?

A It only classifies existing data into categories without creating anything new.

B It creates new content like images, text, or music based on patterns it learned.

C It stores large amounts of data without processing or generating anything.

D It deletes irrelevant data to make storage more efficient.

💡 Hint

Think about how AI can make something new rather than just sorting or storing.

Generative AI learns patterns from data and uses them to create new, original content like text, images, or sounds.

❓ Predict Output

intermediate

Output of a Simple Text Generation Model

What will be the output of this simple text generation code snippet?

Prompt Engineering / GenAI

import random
words = ['cat', 'dog', 'bird']
def generate_sentence():
    return 'I see a ' + random.choice(words)

print(generate_sentence())

A I see a random animal

B I see a cat dog bird

C Error because random.choice is not defined

D I see a cat (or dog or bird) - one of these randomly chosen each run

💡 Hint

Look at how random.choice picks one word from the list each time.

The function picks one word randomly from the list and adds it to the sentence, so output changes each run.

❓ Model Choice

advanced

Choosing the Right Model for Generative AI

Which model type is best suited for generating new images based on learned patterns?

A Generative Adversarial Network (GAN) designed to create new images

B Convolutional Neural Network (CNN) used for image classification

C Recurrent Neural Network (RNN) used for time series prediction

D Decision Tree used for data splitting

💡 Hint

Think about models that can create new images, not just analyze them.

GANs have two parts that compete to generate realistic new images, making them ideal for image generation.

❓ Metrics

advanced

Evaluating Generative AI Quality

Which metric is commonly used to measure how realistic images generated by AI are?

A Inception Score measuring image quality and diversity

B Accuracy score comparing predicted labels

C Mean Squared Error measuring numeric prediction error

D Confusion Matrix showing classification errors

💡 Hint

This metric looks at both how good and how varied generated images are.

Inception Score evaluates generated images by checking if they look like real images and are diverse.

🔧 Debug

expert

Debugging a Text Generation Model Output

Given this code snippet for generating text, what error will it raise?

Prompt Engineering / GenAI

def generate_text(words):
    return ' '.join(words)

print(generate_text('hello world'))

A SyntaxError due to missing colon

B NameError because 'words' is not defined

C No error, outputs 'h e l l o w o r l d'

D TypeError because 'str' object is not iterable as expected

💡 Hint

Consider whether passing a string to ' '.join() raises a TypeError and what the actual output is.

No error is raised. Strings are iterable character-by-character, so ' '.join('hello world') produces 'h e l l o w o r l d'.